The overall goals of Project 3 are to understand the high-resolution structural correlates of immunodominance and to provide structural foundations for designing immunogens that can focus an immune response on particular, chosen epitopes. Our principal hypothesis is that structural contributions to immunodominance can be revealed by correlating structures of antibody-antigen complexes with knowledge of the repertoire from which the antibody was derived and reconstruction of the mutational sequence that led to the affinity matured antibody from its germline precursor. The high-throughput approaches described in Project 1 can give us the required information about repertoire and about affinity maturation of antibodies denved from the settings of vaccination vs. infection. We propose here to provide the corresponding three-dimensional information. We have the following four Aims. [1) What are the structural characteristics of cross-reactive antibodies bound with their HA epitopes? We will determine structures of selected representatives of the antibodies studied in projects 1 and 2, in complex with the relevant influenza virus HA. We will begin with standard crystallographic approaches, but proceed as rapidly as possible to higher-throughput cryoEM methods (Aim 2;see also Core C). (2) We will develop approaches for mounting "guest proteins" [with influenza HA as the principal target) onto icosahedral-virus scaffolds, to create a symmetric array suitable for structure determination by cryoEM. We will start with the rotavirus double-layer particle (DLP) as our scaffold, and the outer-layer glycoprotein, VP7, as the mounting device.;we will also test the other outer-layer protein, VP4, as a trimeric mounting protein. [3) What makes an epitope immunodominant? The opportunity afforded by the cryoEM approach to examine a systematic and relatively unbiased set of complexes will allow us to determine the structural correlates of immunodominance directly. We will determine structures of a selected HA in complex with antibodies representing the entire spectrum of a typical repertoire as revealed by the studies in Project 1 and mapped by Project 2. [4) Application to other viral antigens. We will apply to the mounting strategies developed in Aim 2 to other viral antigens for which a combination of repertoire analysis and structural dissection might lead to novel